1. Field of the Invention
The present invention relates to an image processing method and an apparatus therefor for recording a color image with dots of different colors.
2. Description of the Prior Art
In case of recording a color image with dots of different colors, for example ink dots of cyan, magenta, yellow and black colors, it is necessary to record intermediate gradations in each color.
For the purpose of recording intermediate gradations in color image recording with color dots, there is already known a method of constituting each pixel with a screen dot pattern composed of a matrix of small dots, and representing the gradation of said pixel by the area and arrangement of high-density small dots in said matrix.
Also in order to represent arbitrary colors with dots of limited colors for example in a color ink jet printer, there is employed a method of recording such dots in overlapping manner to achieve subtractive color mixing. However, in the case of representing intermediate colors by superposing screen dots of different colors for each pixel, unevenness in the recorded color may result if the extent of overlapping of different color dots varies, because the state of subtractive color mixing also varies, for example, due to a change in the internal reflection in each ink dot, thus giving rise to variation in the resulting color. If the same arrangement of screen dots is employed for all colors, the above-mentioned unevenness in color is difficult to avoid since it is extremely difficult to record such screen dots without variation in the extent of overlapping of said dots. Therefore, in order to avoid such difficulty of color unevenness, there is employed a method of arranging the screen dots of different colors respectively in different so-called screen angles, so that exact overlapping of different color dots need not be controlled even when the same dot arrangement is employed for different colors. The use of such screen angles limits such color unevenness, if any, within a small area, thereby rendering such color unevenness inconspicuous.
However, in the conventional method, such a screen angle is obtained by merely inclining the screen dot pattern composed of a square dot matrix. It has therefore been necessary, for recording a color image with screen dots of desired screen angles, to store the binary data for each dot matrix in the screen patterns for different colors with different screen angles in a memory device, for example, a read-only memory, and to read such stored data in response to the gradation signal of each pixel of the color image to be recorded. Therefore such conventional color image processing method has been associated with the serious drawback of requiring an enormous memory capacity.
More specifically, for example, color image recording with screen dots having different screen angles respectively for different colors and each composed of a 4.times.4 square dot matrix as shown in FIG. 5E for representing 16 density levels, there will be required a memory capacity of 4.times.4.times.16=256 bits for each color. Also in the use of a screen dot composed of a 17.times.17 dot matrix as shown in FIG. 5H for recording a color image of a relatively high quality there will be required a memory capacity of 17.times.17.times.16=4,624 bits for each color, and a screen dot composed of a 32.times.32 dot matrix for a still finer image quality requires a memory capacity of 32.times.32.times.16=16,384 bits for each color.